mm/hugetlbfs: fix error handling when setting up mounts
authorMike Kravetz <mike.kravetz@oracle.com>
Sun, 1 Dec 2019 01:56:34 +0000 (17:56 -0800)
committerLinus Torvalds <torvalds@linux-foundation.org>
Sun, 1 Dec 2019 20:59:08 +0000 (12:59 -0800)
It is assumed that the hugetlbfs_vfsmount[] array will contain either a
valid vfsmount pointer or NULL for each hstate after initialization.
Changes made while converting to use fs_context broke this assumption.

While fixing the hugetlbfs_vfsmount issue, it was discovered that
init_hugetlbfs_fs never did correctly clean up when encountering a vfs
mount error.

It was found during code inspection.  A small memory allocation failure
would be the most likely cause of taking a error path with the bug.
This is unlikely to happen as this is early init code.

Link: http://lkml.kernel.org/r/94b6244d-2c24-e269-b12c-e3ba694b242d@oracle.com
Reported-by: Chengguang Xu <cgxu519@mykernel.net>
Fixes: 32021982a324 ("hugetlbfs: Convert to fs_context")
Signed-off-by: Mike Kravetz <mike.kravetz@oracle.com>
Cc: David Howells <dhowells@redhat.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
fs/hugetlbfs/inode.c

index 6e5eade..0cacf99 100644 (file)
@@ -1461,28 +1461,41 @@ static int __init init_hugetlbfs_fs(void)
                                        sizeof(struct hugetlbfs_inode_info),
                                        0, SLAB_ACCOUNT, init_once);
        if (hugetlbfs_inode_cachep == NULL)
-               goto out2;
+               goto out;
 
        error = register_filesystem(&hugetlbfs_fs_type);
        if (error)
-               goto out;
+               goto out_free;
 
+       /* default hstate mount is required */
+       mnt = mount_one_hugetlbfs(&hstates[default_hstate_idx]);
+       if (IS_ERR(mnt)) {
+               error = PTR_ERR(mnt);
+               goto out_unreg;
+       }
+       hugetlbfs_vfsmount[default_hstate_idx] = mnt;
+
+       /* other hstates are optional */
        i = 0;
        for_each_hstate(h) {
+               if (i == default_hstate_idx)
+                       continue;
+
                mnt = mount_one_hugetlbfs(h);
-               if (IS_ERR(mnt) && i == 0) {
-                       error = PTR_ERR(mnt);
-                       goto out;
-               }
-               hugetlbfs_vfsmount[i] = mnt;
+               if (IS_ERR(mnt))
+                       hugetlbfs_vfsmount[i] = NULL;
+               else
+                       hugetlbfs_vfsmount[i] = mnt;
                i++;
        }
 
        return 0;
 
- out:
+ out_unreg:
+       (void)unregister_filesystem(&hugetlbfs_fs_type);
+ out_free:
        kmem_cache_destroy(hugetlbfs_inode_cachep);
- out2:
+ out:
        return error;
 }
 fs_initcall(init_hugetlbfs_fs)